Analyst Credibility: The Investor's Perspective [*].(Brief Article)

RESULTS

Of the 177 subjects participating in the study, 86 received the optimistic analyst forecast and 91 received the pessimistic forecast. Unequal cell sizes are present because the instruments were randomly assigned to investors so that no systematic distribution pattern occurred that might bias the results. Fewer subjects were available than anticipated which, combined with the randomized distribution, resulted in a difference between the number of optimistic and pessimistic forecast observations. A manipulation check at the end of the experiment revealed that no subjects were familiar with the company or were able to guess the correct name of the company. Responses to the dependent and independent variables were not significantly different between the two companies; hence the data were combined for hypothesis testing purposes.

Recall that the dependent measure, analyst credibility, is measured using a four-item scale. Factor analysis was used to aggregate the items into one variable. Factor analysis is preferable to a simple average of the four items because it minimizes the error variance inherent among the items and constructs a more accurate value of the underlying construct (Stewart, 1981; Peter et al., 1993). The output of the factor analysis is a set of factor scores or weights assigned to each of the four items. The factor scores are used to construct a weighted average of the four items that is designed to minimize the unique error variance.

The eigenvalue was greater than 1.0 for the analyst credibility factor, indicating the correlation among the items was sufficient to be considered a single construct. In addition, all factor loadings were greater than 0.50 (the smallest was 0.62768), indicating that all four items could be included in the construction of the construct. Thus, the analyst credibility construct was represented as the factor-weighted average of all four items comprising the scale.

The first independent variable, Expected Optimism, was a single-item measure captured on a Likert-type scale that ranged from 1 to 7 (strongly disagree to strongly agree). Analyst forecast type was a manipulated variable that was represented as either a 1 (optimistic forecast) or a 0 (pessimistic forecast).

Regression analysis was used to determine the relationships among the independent and dependent variables. Table 1 illustrates the statistics from the Analyst Credibility Model. The model was significant (F=9. 162; p = 0.0002). Both Expected Optimism and Analyst Forecast Type were statistically significant variables (p = .0363 and .0004, respectively) in the model. The coefficients were negative, which supports the hypotheses. Hence, higher levels of Expected Optimism resulted in lower credibility assessments. When the investors were presented with an optimistic forecast, on average, they rated the credibility of the analyst lower than did subjects receiving the pessimistic forecast.

The Effect of Credibility Assessments on Forecast Reliance

The control group was employed to determine whether the presence of the additional financial information influenced investors' forecast reliance. There was no significant difference between their forecasts and their forecast revisions. Thus, it was assumed that any significant differences between investors' forecast and forecast revision, in the group receiving the analyst forecast, were attributable to the presence of the analyst forecast.

Table 2 illustrates the statistical results stemming from the second model. The data indicate a significant positive relationship between credibility perceptions and investors' reliance on the analysts' forecasts (T=3.505; p = 0.0006). Investors' perceptions of analyst credibility are one important criterion in their decision to rely upon the analyst's recommendation.

DISCUSSION AND IMPLICATIONS

The results of the experiment demonstrate that investors use the analyst's forecast as one signal of the analyst's credibility. When that forecast is optimistically biased, the credibility of the analyst is perceived to be lower relative to an equally pessimistic forecast bias. Furthermore, it was demonstrated that such credibility assessments directly affect the extent to which investors rely upon the analyst's forecast when making their own predictions. No research is void of trade-offs and thus limitations. In this study, subjects are investors who are members of NAIC, an organization that promotes a unified analysis focus. This choice enhanced the research design by controlling for variance in investment analysis methods. However, when generalizing to a larger population, one must consider the findings in this light. Second, the information available to investors was constrained in order to achieve experimental efficiency. Meticulous attention was focused on including variables recommended by NAIC. Howev er, investors use a broad range of information sources and had a key statistic been absent, their forecasts may differ from those generated in a field setting.

Investors are a diverse constituency and often are not vocal in their demands on analysts. The subtly of the investor constituency is lost among the more blatant demands from management and investment banking. John Keefe from Keefe Worldwide Information Services sums up this notion by saying, "the little guy is watching the game and betting the game, but isn't really in the game." (Woolley, 1996: 119). However, investors may have a collective voice in "the game." This study demonstrates that their reliance upon analysts is partially a function of their assessment of analyst credibility and analyst credibility, in the eyes of sophisticated individual investors, is affected when analysts are optimistic. Without substantial credibility, analysts are unable to perform in their most critical roles as efficient conduits between management and investors (Regan, 1993). Knowledge of the factors that influence investors' assessments of analysts' credibility will help the profession take the necessary steps (e.g., cont inuing education that reinforces the professions' role and each analyst's responsibilities) to insure they continue as a valuable resource to all market participants. Future research in this area is needed to determine additional scenarios and analyst characteristics that enhance or hinder an analyst's credibility.

The findings here have implications for management behavior. Corporate managers need analysts because they help minimize the risk inherent in the agency relationship by offering an independent opinion and forecast of corporate activities. As public information sources, analysts also facilitate efficient dissemination of information relative to stock valuation. When management pressures analysts to issue reports that put the company in favorable light, they put at risk a valuable resource they depend upon to distribute credible forward-looking information. Hence, even in the short term, where the analyst may not hold a favorable opinion of the firm, and especially in the long term, the analyst is a critical, irreplaceable resource. Pressuring analysts to conform to management's perspective can damage analysts' credibility with investors, which ultimately undermines their value to management. Given the central role that analysts play in the financial markets, it is in the interests of management, investment ba nkers and others with the ability to leverage their relationship with analysts to step back and envision the trade-offs that are made when analysts are asked to compromise their professional responsibilities. In the light of the societal role analysts play, most would agree that it is in everyone's interest for analysts to be perceived as independent and objective communicators of company activities. The analysts also have a responsibility to clearly emphasize the importance of adhering to their professional code of conduct (AIMR, 1992). Because the implications are subtle, rather than explicit and verifiable, it is easy to see how overt pressures can lead to compromises. However, as this study demonstrates, investors are keenly aware of these compromises. It affects their perceptions of analysts' actions and ultimately the extent to which they value analysts' contributions to the market.

(*.) I wish to express my appreciation to the participants at the Portland-Vancouver Accounting Research Colloquium (PVARC), especially Claire Latham, the NAIC for their subject support, two anonymous reviewers, and the editor.

References

AIMR. 11992. Standards of Practice Handbook, 6th ed. Charlottesville, VA: AIMR.

Ali, A., A. Klein and J. Rosenfeld. 1992. "Analysts Use Of Information About Permanent And Transitory Components In Forecasting Annual EPS." The Accounting Review 67 (January): 183-198.

Dutta, S. and B. Trueman. 1997. "Optimists, Pessimists, and Managerial Disclosures." Working Paper. University of California, Berkeley.

Epstein, M.J. and M.L. Pava. 1993. The Shareholder's Use of Corporate Annual Reports. Greenwich, CT: JAI Press.

Galant, D. 1990. "The Hazards of Negative Research Reports." Institutional Investor July: 73-80.

Gottschalk, E.C. 1994. "Investment Clubs May Be Mired In Stagnant Stocks." Wall Street Journal July 28: C1.

Hayward, M.L.A. and W. Boeker. 1998. "Power and Conflict of Interest in Professional Firms: Evidence from Investment Banking." Administrative Science Quarterly 43 (1): 1.

Heider, F. 1944. "Social Perception and Phenomenal Causality." Psychological Review 51: 358-374.

Hirst, D.E., L. Koonce and P.J. Simko. 1995. "Investor Reactions to Financial Analysts' Research Reports." Journal of Accounting Research 33(2) (Autumn): 335-352.

Jensen, M.C. and W.H. Meckling. 1976. "Theory of the Finn: Managerial Behavior, Agency Costs and Ownership Structure." Journal of Financial Economics 3 (October): 305-360.

Print Get the Mag Weekly Updates